{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:ARM4YTP7VHQDA6TQ7FTZFB7RY3","short_pith_number":"pith:ARM4YTP7","canonical_record":{"source":{"id":"2505.23477","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-29T14:27:14Z","cross_cats_sorted":[],"title_canon_sha256":"35c45e86c6adae7b16721a99a26f59953a163d5ae719fd680bfd29bf699f13a1","abstract_canon_sha256":"a0715eeb1bdd563e7f50f0dae19417083452f3ac26eb1e7761fc02626e3b75dd"},"schema_version":"1.0"},"canonical_sha256":"0459cc4dffa9e0307a70f9679287f1c6ff840855261cb516c4ca289d1046c37b","source":{"kind":"arxiv","id":"2505.23477","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.23477","created_at":"2026-07-05T11:12:02Z"},{"alias_kind":"arxiv_version","alias_value":"2505.23477v1","created_at":"2026-07-05T11:12:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.23477","created_at":"2026-07-05T11:12:02Z"},{"alias_kind":"pith_short_12","alias_value":"ARM4YTP7VHQD","created_at":"2026-07-05T11:12:02Z"},{"alias_kind":"pith_short_16","alias_value":"ARM4YTP7VHQDA6TQ","created_at":"2026-07-05T11:12:02Z"},{"alias_kind":"pith_short_8","alias_value":"ARM4YTP7","created_at":"2026-07-05T11:12:02Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:ARM4YTP7VHQDA6TQ7FTZFB7RY3","target":"record","payload":{"canonical_record":{"source":{"id":"2505.23477","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-29T14:27:14Z","cross_cats_sorted":[],"title_canon_sha256":"35c45e86c6adae7b16721a99a26f59953a163d5ae719fd680bfd29bf699f13a1","abstract_canon_sha256":"a0715eeb1bdd563e7f50f0dae19417083452f3ac26eb1e7761fc02626e3b75dd"},"schema_version":"1.0"},"canonical_sha256":"0459cc4dffa9e0307a70f9679287f1c6ff840855261cb516c4ca289d1046c37b","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:12:02.533214Z","signature_b64":"d2GI7JislAhPuC/n40CEPsfe3qmRnTIwJKgpOkPqB/BlaA3NsBviROMmrvmHF/eB35uWyTb/8POVx3bDALVoCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0459cc4dffa9e0307a70f9679287f1c6ff840855261cb516c4ca289d1046c37b","last_reissued_at":"2026-07-05T11:12:02.531904Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:12:02.531904Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.23477","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:12:02Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"8yNkVrG5JcWNexKng8Dh4sIR86StmPevkxnicdZTSTK16fjmTtqLvmA9nhcEyLSQgizML/uU9XqBQEQnRkKCBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T20:18:19.174280Z"},"content_sha256":"9ce75e02b7edbec8c5fe8be67377ec32385f3b55822512150ec72dbaa562d1dc","schema_version":"1.0","event_id":"sha256:9ce75e02b7edbec8c5fe8be67377ec32385f3b55822512150ec72dbaa562d1dc"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:ARM4YTP7VHQDA6TQ7FTZFB7RY3","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Evaluating the performance and fragility of large language models on the self-assessment for neurological surgeons","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Anton Alyakin, Daniel Alexander Alber, Douglas Kondziolka, Eric Karl Oermann, Jin Vivian Lee, Karl L. Sangwon, Krithik Vishwanath, Mrigayu Ghosh","submitted_at":"2025-05-29T14:27:14Z","abstract_excerpt":"The Congress of Neurological Surgeons Self-Assessment for Neurological Surgeons (CNS-SANS) questions are widely used by neurosurgical residents to prepare for written board examinations. Recently, these questions have also served as benchmarks for evaluating large language models' (LLMs) neurosurgical knowledge. This study aims to assess the performance of state-of-the-art LLMs on neurosurgery board-like questions and to evaluate their robustness to the inclusion of distractor statements. A comprehensive evaluation was conducted using 28 large language models. These models were tested on 2,904"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.23477","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2505.23477/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:12:02Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7R4Kb3WT2+KOgrNO60cOBhWA/cerk4Jh6q7ngIz/DjWZRiIRw67ZRnWXkmNXSqKWk3SCVKvRZ7ld0YrNC4cqDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T20:18:19.174662Z"},"content_sha256":"db2ee79632c6f434828361c279edabb280a7c0566c54f05f809925b51cd9b50a","schema_version":"1.0","event_id":"sha256:db2ee79632c6f434828361c279edabb280a7c0566c54f05f809925b51cd9b50a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ARM4YTP7VHQDA6TQ7FTZFB7RY3/bundle.json","state_url":"https://pith.science/pith/ARM4YTP7VHQDA6TQ7FTZFB7RY3/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ARM4YTP7VHQDA6TQ7FTZFB7RY3/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-07-06T20:18:19Z","links":{"resolver":"https://pith.science/pith/ARM4YTP7VHQDA6TQ7FTZFB7RY3","bundle":"https://pith.science/pith/ARM4YTP7VHQDA6TQ7FTZFB7RY3/bundle.json","state":"https://pith.science/pith/ARM4YTP7VHQDA6TQ7FTZFB7RY3/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ARM4YTP7VHQDA6TQ7FTZFB7RY3/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:ARM4YTP7VHQDA6TQ7FTZFB7RY3","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"a0715eeb1bdd563e7f50f0dae19417083452f3ac26eb1e7761fc02626e3b75dd","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-29T14:27:14Z","title_canon_sha256":"35c45e86c6adae7b16721a99a26f59953a163d5ae719fd680bfd29bf699f13a1"},"schema_version":"1.0","source":{"id":"2505.23477","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.23477","created_at":"2026-07-05T11:12:02Z"},{"alias_kind":"arxiv_version","alias_value":"2505.23477v1","created_at":"2026-07-05T11:12:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.23477","created_at":"2026-07-05T11:12:02Z"},{"alias_kind":"pith_short_12","alias_value":"ARM4YTP7VHQD","created_at":"2026-07-05T11:12:02Z"},{"alias_kind":"pith_short_16","alias_value":"ARM4YTP7VHQDA6TQ","created_at":"2026-07-05T11:12:02Z"},{"alias_kind":"pith_short_8","alias_value":"ARM4YTP7","created_at":"2026-07-05T11:12:02Z"}],"graph_snapshots":[{"event_id":"sha256:db2ee79632c6f434828361c279edabb280a7c0566c54f05f809925b51cd9b50a","target":"graph","created_at":"2026-07-05T11:12:02Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2505.23477/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The Congress of Neurological Surgeons Self-Assessment for Neurological Surgeons (CNS-SANS) questions are widely used by neurosurgical residents to prepare for written board examinations. Recently, these questions have also served as benchmarks for evaluating large language models' (LLMs) neurosurgical knowledge. This study aims to assess the performance of state-of-the-art LLMs on neurosurgery board-like questions and to evaluate their robustness to the inclusion of distractor statements. A comprehensive evaluation was conducted using 28 large language models. These models were tested on 2,904","authors_text":"Anton Alyakin, Daniel Alexander Alber, Douglas Kondziolka, Eric Karl Oermann, Jin Vivian Lee, Karl L. Sangwon, Krithik Vishwanath, Mrigayu Ghosh","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-29T14:27:14Z","title":"Evaluating the performance and fragility of large language models on the self-assessment for neurological surgeons"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.23477","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:9ce75e02b7edbec8c5fe8be67377ec32385f3b55822512150ec72dbaa562d1dc","target":"record","created_at":"2026-07-05T11:12:02Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"a0715eeb1bdd563e7f50f0dae19417083452f3ac26eb1e7761fc02626e3b75dd","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-29T14:27:14Z","title_canon_sha256":"35c45e86c6adae7b16721a99a26f59953a163d5ae719fd680bfd29bf699f13a1"},"schema_version":"1.0","source":{"id":"2505.23477","kind":"arxiv","version":1}},"canonical_sha256":"0459cc4dffa9e0307a70f9679287f1c6ff840855261cb516c4ca289d1046c37b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0459cc4dffa9e0307a70f9679287f1c6ff840855261cb516c4ca289d1046c37b","first_computed_at":"2026-07-05T11:12:02.531904Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:12:02.531904Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"d2GI7JislAhPuC/n40CEPsfe3qmRnTIwJKgpOkPqB/BlaA3NsBviROMmrvmHF/eB35uWyTb/8POVx3bDALVoCA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:12:02.533214Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.23477","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9ce75e02b7edbec8c5fe8be67377ec32385f3b55822512150ec72dbaa562d1dc","sha256:db2ee79632c6f434828361c279edabb280a7c0566c54f05f809925b51cd9b50a"],"state_sha256":"e62a2a363efa29cb018bc81bb6d2e5ec26bef2e6168d4cae45be318705e3a2eb"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7pY2y77fp9JCJ0ac6+Ipy3NDPqHXMSs8iieqESpUEdDv6MlD0KcygUJ9r3FjCu/sEHpB7mbWExs6k0HM2saGAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-06T20:18:19.176651Z","bundle_sha256":"c95e98717ea0caef29f433531ebc588d8f28323a2da33820559293b4b6cb09f4"}}